Consolidate Compass skill entrypoints

AGENTS.md

@@ -67,8 +67,8 @@ user has already granted or explicitly waived that review boundary.
 
 ## Skill Authoring
 
-Use `write-a-skill` for general skill design and `write-a-compass-skill` for
-Compass routing, install, derivation, provenance, retirement, and validation.
+Use `write-a-skill` for skill design. Use `compass` to adopt the result into
+reviewed Compass source.
 
 A skill should make the reader absorb, in its first screen:
 

codex/skills/compass/SKILL.md

@@ -1,57 +1,34 @@
 ---
 name: compass
-description: Maintain Compass source, runtime boundaries, install wiring, retirement, and validation. Invoke manually for Compass repository changes.
+description: Capture or apply durable learning in Compass, or refresh the live setup. Use when asked to write, save, adopt, update, or sync something through Compass.
 ---
 
 # Compass
 
-Maintain Compass as reviewed source for portable Codex and Claude behavior. This
-skill exists because a strong runtime change can still fail when source routing,
-derivation, retirement, policy, and verification disagree.
+Turn the user's learning or requested change into the smallest coherent Compass
+update. Use this as the single entrypoint for changing reviewed Compass source or
+refreshing the live setup.
 
-Use it only when changing Compass itself.
+Inspect the Compass repository before deciding where the change belongs. Treat
+its current repository guidance, workflows, manifests, scripts, and tests as
+authoritative. Do not restate or cache their rules here.
 
-## Preserve The Source Boundaries
+## Apply The Learning
 
-- Codex global guidance belongs in `codex/AGENTS.md`.
-- Claude global guidance belongs in `claude/CLAUDE.md`.
-- Runtime-neutral shared skills and agents belong under `codex/` and derive only
-  where the manifest says so.
-- Claude-specific agent contracts belong under `claude/agents/`.
-- Portable opt-in domain packs belong under `carried/` and stay out of global
-  install lists.
-- Compass maintainer reasoning belongs in `workflows/` and `local-docs/`.
-- Deterministic truth belongs in scripts, hooks, manifests, schemas, and tests.
+Find the existing surface that owns the behavior. Prefer refining, merging,
+moving, mechanizing, or retiring material over adding another source.
 
-Read root `AGENTS.md`, `philosophy.md`, `workflows/addition-intake.md`,
-`local-docs/maintenance-learnings.md`, and the workflow nearest the change.
+When the change creates or revises a reusable skill, use `write-a-skill` for the
+skill's contract and language, then follow the repository's current integration
+route.
 
-## Change The Whole Ownership Contract
+When the request is only to refresh the live installation, use the repository's
+reviewed update workflow and script instead of changing source.
 
-When behavior or ownership moves, update the complete route in one change:
+## Result
 
-- reviewed source;
-- Codex and Claude install or derivation maps;
-- source provenance;
-- explicit retirement paths;
-- policy and required-file checks;
-- MCP catalog expectations;
-- narrow tests and install round-trip coverage;
-- directly related documentation.
+Finish with one authoritative expression of the learning, the smallest coherent
+diff, directly affected wiring reconciled, stale or duplicate paths removed,
+current verification evidence, and a focused pull request when source changed.
 
-Make Compass-owned capability exact. A fallback paragraph is not a substitute
-for correct wiring.
-
-## Validate The Result
-
-Run narrow tests first, then:
-
-```powershell
-.\scripts\doctor.ps1
-.\scripts\verify-live.ps1 -SkipCodexCommand
-git diff --check
-```
-
-Exercise install and retirement paths when ownership changes. Forward-test
-judgment changes with a fresh realistic invocation. Use a focused PR as the
-review unit and keep public mutation behind named authority.
+Do not merge or make another public mutation without named authority.

codex/skills/compass/agents/openai.yaml

@@ -1,7 +1,7 @@
 interface:
   display_name: "Compass"
-  short_description: "Maintain portable Compass setup"
-  default_prompt: "Use $compass to update durable Compass source and install wiring."
+  short_description: "Capture durable learning in Compass"
+  default_prompt: "Use $compass to write this learning into the smallest authoritative Compass surface, or refresh the live setup when that is the request."
 
 policy:
   allow_implicit_invocation: false

codex/skills/write-a-skill/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: write-a-skill
-description: Create or revise a reusable agent skill that strongly orients judgment, stays lean, and proves its behavior.
+description: Create, revise, or retire a reusable agent skill that strongly orients judgment, stays lean, and preserves a clean lifecycle.
 ---
 
 # Write A Skill
@@ -78,6 +78,17 @@ contract must stay exact.
 Write decisively. Succinct and pithy language helps only when the role remains
 complete.
 
+## Preserve One Lifecycle
+
+Maintain one canonical source and derive copies mechanically. Do not hand-edit
+installed or generated copies.
+
+When adapting an external skill, preserve its source, immutable version, license,
+and reviewed changes outside normal runtime guidance.
+
+A rename is retirement plus a new skill. Remove stale names, triggers, install
+entries, and owned live copies without touching unrelated user material.
+
 ## Prove The Skill
 
 Review behavior, not prose alone:
@@ -88,11 +99,12 @@ Review behavior, not prose alone:
 - Can detail move to a reference or deterministic mechanic?
 - Does any rule decide something that should remain contextual?
 - Can any paragraph disappear without losing behavior?
+- Do source, derived copies, discovery metadata, and retirement state agree?
 
 Forward-test fragile judgment with realistic prompts. Repair the cause of weak
 behavior instead of appending another warning.
 
 ## Output
 
 Return the skill, needed resources, trigger rationale, behavior evidence, known
-boundaries, and the durable surface that owns it.
+boundaries, lifecycle changes, and the durable surface that owns it.

codex/skills/write-a-skill/agents/openai.yaml

@@ -1,7 +1,7 @@
 interface:
   display_name: "Write A Skill"
-  short_description: "Write strong lean reusable agent skills"
-  default_prompt: "Use $write-a-skill to create or revise a reusable skill that establishes a strong role, desired result, evidence standard, and authority boundary before procedure."
+  short_description: "Write and maintain lean agent skills"
+  default_prompt: "Use $write-a-skill to create, revise, or retire a reusable skill whose role, desired result, evidence standard, authority boundary, and lifecycle are clear before procedure."
 
 policy:
   allow_implicit_invocation: false

manifests/policy-contracts.json

@@ -59,17 +59,14 @@
       }
     },
     {
-      "id": "routing-source-referenced-by-compass-skill",
+      "id": "compass-single-entrypoint",
       "path": "codex/skills/compass/SKILL.md",
       "required": [
-        "workflows/addition-intake.md"
-      ]
-    },
-    {
-      "id": "routing-source-referenced-by-skill-authoring",
-      "path": "codex/skills/write-a-compass-skill/SKILL.md",
-      "required": [
-        "workflows/addition-intake.md"
+        "Turn the user's learning or requested change into the smallest coherent Compass",
+        "current repository guidance, workflows, manifests, scripts, and tests as",
+        "Do not restate or cache their rules here.",
+        "use `write-a-skill`",
+        "reviewed update workflow and script instead of changing source."
       ]
     },
     {

manifests/portable-files.toml

@@ -37,11 +37,9 @@ skills = [
   "specialist-review",
   "subagent-driven-development",
   "to-prd",
-  "update-compass",
   "using-goals",
   "which-llm",
   "workspace-steward",
-  "write-a-compass-skill",
   "write-a-skill",
 ]
 
@@ -82,7 +80,6 @@ derived_skills = [
   "to-prd",
   "using-goals",
   "workspace-steward",
-  "write-a-compass-skill",
   "write-a-skill",
 ]
 

manifests/retired-skills.json

@@ -16,16 +16,19 @@
     "input-token-economy",
     "pr-merge-closeout",
     "ui-ux-pro-max",
+    "update-compass",
     "using-codex-goals",
     "webmcp-eval-triage",
     "webmcp-tool-authoring",
-    "webmcp-verify-tool"
+    "webmcp-verify-tool",
+    "write-a-compass-skill"
   ],
   "claude_skills": [
     "benchmark-infra-reviewer",
     "benchmark-run-operator",
     "input-token-economy",
-    "using-codex-goals"
+    "using-codex-goals",
+    "write-a-compass-skill"
   ],
   "claude_agents": [
     "benchmark-infra-reviewer.md"

manifests/skill-sources.json

@@ -146,16 +146,6 @@
         "claude": "derived"
       }
     },
-    {
-      "name": "update-compass",
-      "owner": "compass",
-      "source": "codex/skills/update-compass",
-      "profile": "default",
-      "targets": {
-        "codex": "copy",
-        "claude": "none"
-      }
-    },
     {
       "name": "using-goals",
       "owner": "compass",
@@ -191,16 +181,6 @@
         "claude": "derived"
       }
     },
-    {
-      "name": "write-a-compass-skill",
-      "owner": "compass",
-      "source": "codex/skills/write-a-compass-skill",
-      "profile": "default",
-      "targets": {
-        "codex": "copy",
-        "claude": "derived"
-      }
-    },
     {
       "name": "write-a-skill",
       "owner": "compass",